CiAO: An Aspect-Oriented OS Family for Resource-Constrained Embedded Systems Daniel Lohmann Wanja Hofer Wolfgang Schröder-Preikschat Friedrich-Alexander University Erlangen-Nuremberg Jochen Streicher Olaf Spinczyk Technical University Dortmund CiAO is aspect-oriented
Motivation: Embedded Systems 2 2 wosch@cs.fau.de - USENIX 2009
Motivation: Embedded Systems Goal: Reuse Scalability, Configurability, Tailorability 3 3 wosch@cs.fau.de - USENIX 2009
Motivation: Configurability of eCos 4 4 wosch@cs.fau.de - USENIX 2009
Motivation: Configurability of eCos Mutex options: PROTOCOL CEILING INHERIT DYNAMIC Kernel policies: Tracing Instrumentation Synchronization 5 5 wosch@cs.fau.de - USENIX 2009
Motivation: Configurability of eCos Mutex options: PROTOCOL CEILING INHERIT DYNAMIC Kernel policies: Tracing Instrumentation Synchronization 6 6 wosch@cs.fau.de - USENIX 2009
Motivation: Configurability of eCos Mutex options: PROTOCOL CEILING INHERIT DYNAMIC Kernel policies: Tracing Instrumentation Synchronization 7 7 wosch@cs.fau.de - USENIX 2009
Motivation: Configurability of eCos Base implementation: Cyg_Mutex::Cyg_Mutex() { locked = false; owner = NULL; } Cross-cutting concerns: 34 #ifdef-blocks spread over 17 functions and data structures in 4 implementation units 8 8 wosch@cs.fau.de - USENIX 2009
Motivation: Configurability of OS #ifdef hell Difficult to understand, maintain, evolve Lack of encapsulation in the implementation Tangling of many concerns in one implementation unit Scattering across several implementation units 11 11 wosch@cs.fau.de - USENIX 2009
Talk Outline Cross-cutting in Configurable System Software Aspect-Oriented Programming (AOP) to the Rescue? CiAO: Aspect-Aware OS Design Evaluation: Suitability for Embedded Systems 12 12 wosch@cs.fau.de - USENIX 2009
Talk Outline Cross-cutting in Configurable System Software Aspect-Oriented Programming (AOP) to the Rescue? CiAO: Aspect-Aware OS Design Evaluation: Suitability for Embedded Systems 13 13 wosch@cs.fau.de - USENIX 2009
AOP to the Rescue? AOP aids modularization of cross-cutting concerns well modularized concern badly modularized without AOP 14 14 wosch@cs.fau.de - USENIX 2009
AOP to the Rescue? AOP aids modularization of cross-cutting concerns It does so by means of aspects aspect well modularized concern badly modularized without AOP with AOP 15 15 wosch@cs.fau.de - USENIX 2009
AOP – Short Introduction Encapsulation of (cross-cutting) concerns in aspects aspect components 16 16 wosch@cs.fau.de - USENIX 2009
AOP – Short Introduction Encapsulation of (cross-cutting) concerns in aspects Aspects give advice to join points in the target system aspect components 17 17 wosch@cs.fau.de - USENIX 2009
AOP – Short Introduction Encapsulation of (cross-cutting) concerns in aspects Aspects give advice to join points in the target system Set of join points described by a pointcut expression aspect pointcut components 18 18 wosch@cs.fau.de - USENIX 2009
AOP – Short Introduction Encapsulation of (cross-cutting) concerns in aspects Aspects give advice to join points in the target system Set of join points described by a pointcut expression aspect pointcut Obliviousness & Quantification components 19 19 wosch@cs.fau.de - USENIX 2009
AOP – AspectC++ Extension to C++ C++ Sources AspectC++ Sources Source-to-source weaver ac++ Woven C++ Code c++ http://www.aspectc.org 01010001100 11011101001 Binary Program 20 20 wosch@cs.fau.de - USENIX 2009
AOP – Why Can It Help? Inversion of caller–callee binding: // traditional implementation void dingus() { …// do basic stuff #ifdef OPT_FEATURE thingy(); #endif } void thingy() { …// do optional stuff } 21 21 wosch@cs.fau.de - USENIX 2009
AOP – Why Can It Help? Inversion of caller–callee binding: // traditional implementation // AOP implementation void dingus() void dingus() { { …// do basic stuff // do basic stuff #ifdef OPT_FEATURE } thingy(); #endif } void thingy() advice execution(“% dingus()”) : { after() …// do optional stuff { } // do optional stuff } 22 22 wosch@cs.fau.de - USENIX 2009
AOP – Why Can It Help? Quantification over multiple join points: // traditional implementation void dingus_foo() { #ifdef OPT_FEATURE thingy(); #endif } void dingus_bar() { #ifdef OPT_FEATURE thingy(); #endif } void thingy(){ // do optional stuff } 23 23 wosch@cs.fau.de - USENIX 2009
AOP – Why Can It Help? Quantification over multiple join points: // traditional implementation // AOP implementation void dingus_foo() { void dingus_foo() { #ifdef OPT_FEATURE } thingy(); #endif } void dingus_bar() { void dingus_bar() { #ifdef OPT_FEATURE } thingy(); #endif } void thingy(){ advice execution(“% dingus % ()”) : // do optional stuff after() { } // do optional stuff } 24 24 wosch@cs.fau.de - USENIX 2009
Talk Outline Cross-cutting in Configurable System Software Aspect-Oriented Programming (AOP) to the Rescue? CiAO: Aspect-Aware OS Design Evaluation: Suitability for Embedded Systems 25 25 wosch@cs.fau.de - USENIX 2009
CiAO: Aspect-Aware OS Design Goal of the CiAO project: Evaluate if AOP is suitable for the design of configurable embedded system software Can AOP help to avoid #ifdef hell? Is AOP efficient enough for the domain of embedded systems? 26 26 wosch@cs.fau.de - USENIX 2009
CiAO: Aspect-Aware OS Design Goal of the CiAO project: Evaluate if AOP is suitable for the design of configurable embedded system software Can AOP help to avoid #ifdef hell? Is AOP efficient enough for the domain of embedded systems? Yes , if the system is designed in an aspect-aware manner Obliviousness? No! Quantification? Yes! 27 27 wosch@cs.fau.de - USENIX 2009
CiAO: Aspect-Aware OS Design Idea of aspect awareness: Provide unambiguous and statically evaluable join points. Unambiguity : Important system state transitions can be captured by a pointcut expression. Static evaluation : Avoid necessity for dynamic pointcut functions, which bear an overhead. 28 28 wosch@cs.fau.de - USENIX 2009
CiAO: Aspect-Aware OS Design Loose coupling Visible transitions Minimal extensions Result: Sparse base system designed with classes, most functionality provided by (optional) aspects Three main aspect purposes: Extension aspects : add features Policy aspects : glue components together Upcall aspects : bind behavior to lower-level events 31 31 wosch@cs.fau.de - USENIX 2009
Extension Aspects Example: task scheduling // base implementation struct Task { Pri pri_; State state_; … }; class Sched { Tasklist ready_; Task::Id running_; public: void activate(Task::Id t); void reschedule(); … }; 32 32 wosch@cs.fau.de - USENIX 2009
Extension Aspects Example: task scheduling extended (resource control) // base implementation aspect ResourceSupport { struct Task { advice “Task” : slice struct { Pri pri_; ResourceMask occupied_; State state_; Pri originalPri_; … }; }; advice “Sched” : slice class { class Sched { public: Tasklist ready_; void getRes(Res::Id r) { Task::Id running_; // lock mutex public: } void activate(Task::Id t); void relRes(Res::Id r) { void reschedule(); // unlock mutex … } }; }; }; ResourceSupport.ah 33 33 wosch@cs.fau.de - USENIX 2009
Policy Aspects Example: specification of preemption points aspect FullPreemption { // points where another task may get a higher prio pointcut pcPreemptionPoints = “% Sched::activate(…)” || “% Sched::setEvent(…)” || “% Sched::relRes(…)”; advice execution(pcPreemptionPoints()) : after() { tjp->that()->reschedule(); } }; Preemption.ah 34 34 wosch@cs.fau.de - USENIX 2009
Upcall Aspects Example: binding of an interrupt handler aspect TimerBinding { advice execution(“% irq::Timer::handler(…)”) : after() { // handle IRQ } }; TimerBinding.ah 35 35 wosch@cs.fau.de - USENIX 2009
Talk Outline Cross-cutting in Configurable System Software Aspect-Oriented Programming (AOP) to the Rescue? CiAO: Aspect-Aware OS Design Evaluation: Suitability for Embedded Systems 36 36 wosch@cs.fau.de - USENIX 2009
Evaluation Suitability of AOP for configurable systems Increased modularization? Scalability? Suitability of AOP for resource-constrained systems Resource efficiency? Scalability? 37 37 wosch@cs.fau.de - USENIX 2009
Evaluation Suitability of AOP for configurable systems Increased modularization? Scalability? Suitability of AOP for resource-constrained systems Resource efficiency? Scalability? 38 38 wosch@cs.fau.de - USENIX 2009
Evaluation Suitability of AOP for configurable systems Increased modularization? Scalability? Suitability of AOP for resource-constrained systems Resource efficiency? Scalability? 39 39 wosch@cs.fau.de - USENIX 2009
Recommend
More recommend
